This proposal will be used to train the applicant in the positional cloning of important genomic regions in human brain tumors. Both a didactic portion, encompassing courses in molecular biology and statistics, and research training, using Targe fragment clones, are included. Currently, the most efficient way to isolate genes in specific chromosomal regions (positional cloning) is to manipulate large genomic clones, such as sosmides and yeast artificial chromosomes. By (1) constructing, screening, and sequencing the ends of cosmic genomic clones, (2) isolating yeast artificial chromosome (YAC) clones, and (3) mapping genomic markers, the applicant will become facile in the techniques necessary to clone chromosomal regions and genes relevant in human brain tumors. In addition, by utilizing published genomic databases in this proposal, the applicant will become able to manipulate these resources, which will be extremely important for work in the future. This proposal will complement the applicant~s previous training in molecular cytogenetics and cDNA manipulation, and provide experience necessary for the transition to an independent investigator. The incidence of primary brain neoplasms in the United States is approximately 20,000 cases/years, which is rising. Despite the relatively low incidence, these tumors are the second leading cause of cancer related death in children, the third between the ages of 15-34 years, and the fourth from ages 35-45 years. Primary brain neoplasm thus represents a deadly tumor for which the prognosis is grim. The majority of these tumors are of glial origin, with the most frequently diagnosed tumor being the glioblastoma. These tumors are thought to be the result of sequential genetic changes, allowing unregulated growth and changes in differentiation. In gliomas, an ontologically intermediate genetic event, gene amplification, has been found to be extraordinarily frequent (up to 50% of cases) in higher grade tumors. Indeed, because of its frequency, as well as the nature of genes identifies, gene amplification has been thought to be particularly relevant in these tumors. However, the knowledge of genes amplified in gliomas is incomplete. Recent molecular cytogenetic studies have revealed diverse chromosomal regions which show increases in copy number which have not been previously associated with these primary brain neoplasms. These regions are expected to contain gene(s) which are importanat in the pathogenesis of these tumors. A particularly interesting region lies on chromosome 20g, which shows increased copy munber in up to 80% of high grade glioma, and in additon, 40% of breast carcinoma cell lines. As a result, because of this frequentcy and diversity, the applicant hypothesizes this region contains gene(s) which are important in both glial and breast tumor genesis, and may have more globel importance in cencer. This proposal is to define the extent of this amplified chronosome 20 regions to determine the 'Core amplicon' (or amplification unit), which encompasses the region consistently amplifiedin gliomas. These studies are imperative and a prerequisite prior to cloning and obtaining the target genes of the region. The approach to be used will be to (1) isolate region specific clones by chromosome microdissection; 92) obtain larger fragment clones (cosmids) to highly represent the regon; and (3) identify a yearst artificial chromosome clone, as an ~anchor~ clone. Somce Yacs typically hae associated known markers, and because as part of the Human Oenome project each chromosome has a genetic linkage map of ordered markers, by obtaining a yeast artificial chromosome shich is part of an amplification unit, one has placed the amplicon into a genetic map. The marker from the identifies YAC ('anchoir narjer') and flanking markers from the anchor marker can then be assessed for amolification to define the extent of the amplivon. In this way by utilizing available genemics resources, the amplificatio unit can be mapped, and future studies can suethese defined ents to identify the large fragment clones from the region' and subsequently candidate genes, as the target for the amplification which ae biologically important in the tumor.